© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 6-RES-JB
    11. 20387
    Prev
    Next
    INDEX
    Explanations

    instances of the word "cannot" with a high activation value

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "cannot" or its variants, indicating denial or inability

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ItemThumbnailImage
    -0.74
    soDeliveryDate
    -0.67
    grounds
    -0.66
    bats
    -0.66
     Generation
    -0.64
    ItemImage
    -0.64
     behavi
    -0.63
    senal
    -0.63
    estate
    -0.62
    Kit
    -0.61
    POSITIVE LOGITS
     afford
    1.18
     necessarily
    1.01
    adian
    0.99
     reproduce
    0.94
     tolerate
    0.93
    't
    0.90
     feas
    0.86
     comprehend
    0.86
     distinguish
    0.84
    icably
    0.84
    Activations Density 0.033%

    No Known Activations